Thanks for looking. The most useful contribution is a clear issue.
- Bug reports: a script gives a wrong result, crashes, or a skill instruction contradicts another one.
- False positives and misses: a check flags something that is fine, or misses something it claims to catch. A small HTML sample that shows it is ideal.
- Ideas: a check, an edition or a document you think is missing.
Pull requests are not accepted at the moment. Open an issue instead and describe the change. This keeps every change going through the same checks before it ships, and it will be revisited as the project settles.
- What you ran (the command, or what you asked your agent to do) and on which edition.
- What you expected and what happened.
- The smallest page or file that reproduces it. Use made-up content.
Never include credentials, customer data, or the name or address of a site you do not own.
Do not open a public issue. See SECURITY.md.
The project is MIT licensed, so you are free to fork it. To run the checks:
python3 -m venv .venv && .venv/bin/pip install -r skills/qa-site-audit/scripts/requirements.txt -r skills/qa-standards/scripts/requirements.txt
(cd tests/node && npm ci)
tools/check.shtools/check.sh runs every test suite, the package build scan and the link check, and
prints one line per check. See docs/maintaining.md for the layout.